Machine vision has become an essential element of quality assurance and process control in manufacturing and its popularity is set to grow.
Advances in hardware, processing power and software algorithms over recent years have allowed companies to automate many tasks that would have been unfeasible only a decade ago. Getting such applications to work in a reliable and cost-effective manner requires considerable skill and experience on the part of the system integrator; frequently, decisions about the lighting, product presentation, camera fixturing and operation of a machine vision system can have as much of an impact on its performance as the choice of appropriate hardware and analysis technology. So where to start...?
Product and process characteristics – all shapes and sizes
Significant variation in the size or shape of the products being inspected by the system can create problems for a single fixed camera position. Similarly, a variation in product colour or surface finish can create challenges for the selection of appropriate lighting. In response to these issues, automatic or manually adjustable camera fixtures can be used to ensure that all the relevant parts of products are in the image, and in focus, while lighting systems are available which automatically adapt to changes in product appearance in order to maintain image quality.
Sometimes there is significant natural variation in the appearance of good products. For example flexible packets, can vary significantly between one product and another. Advanced software approaches, including sophisticated calibration, pattern unwrapping and adaptive tools can be used to overcome these issues.
Appropriate product presentation and precise fixturing can also simplify image acquisition and processing. In general, time spent optimising the image before the processing stage will be repaid many times in the life of the project from a software development, inspection robustness, and system maintenance requirement point of view.
Image requirements – picture resolution
One of the key factors in determining the architecture of an automated vision system is the pixel resolution needed to achieve the required inspection functions. In industrial applications the tightest measured tolerance must typically represent 5 to 10 pixels of the acquired image. So a tolerance of +/- 0.5mm may require a pixel resolution of 100µm.
In addition, any feature to be detected must occupy a number of pixels: single pixel features are subject to noise and ‘edge effects’ and cannot be reliably detected. A good rule of thumb is that a feature should be 3x3 pixels for detection, so a resolution of around 150µm is required to detect features 0.5mm in size. Some special processing tools also have their own requirements. Optical character recognition tools typically require individual characters to be 20 to 30 pixels high, for example, so a 12 point typeface would require a resolution of around 200µm.
Once the resolution is known, it is possible to define the camera and lens combination that will allow for this to be achieved over the object to be inspected. If this leads to very large image requirements, integrators can use multiple cameras, custom optics to select areas of interest, software to select areas of interest, or the use of linescan or contact image sensor (CIS) technology.
Speed – faster the better
If the product is moving continuously then the acquisition must ‘freeze’ the movement to avoid ‘motion blur’ in the image. This can be done through the use of very short exposure times, or with strobe lighting. In both cases intense light is required, and specialised sources are often needed to achieve an adequately bright image. Once the sensor has been exposed, the data must be transferred from the camera to the processor.
In general, high-resolution cameras have lower maximum frame rates, and this is also affected by the data transfer interface. 5 to 100 frames per second are typical in the field. Recently, a number of high-speed camera interfaces have become available, such as GigE.
The time required to analyse images after processing is highly dependent on image content and the algorithms in use. Higher speeds and more complex analyses are facilitated by increased processing power, and the most advanced systems make use of high powered intelligent cameras, and multiple, multi-core PCs with image processing distributed across them.
The bigger picture
Finally, whatever the technologies involved, the machine vision system must work smoothly with the organisation’s wider production and quality assurance processes. In the past, integrating machine vision systems in this way required extensive and labour-intensive custom programming, but today the availability of dedicated integration packages – like Optimal’s synTI system - have greatly simplified, accelerated and reduced the cost of such efforts, ensuring that machine vision is seamlessly integrated into the bigger factory automation picture.
You can contact Optimal to discuss any vision system requirement, plus a huge range of machine building and automation software/hardware integration service, either by email satels@optimal-ltd.co.uk or by phone +44(0) 1454 333 222.
Photo caption:
Picture 1: Cognex InSight camera
Picture 2: Specialist lighting is needed for high speed vision
Picture 3: Geoff Norwood, Applications Engineer, Optimal Industrial Automation
Picture 4: Sealed blister inspection